Development of Algorithms to Convert Large Streams of Truck GPS Data into Truck Trips

Abstract

This paper documents the development and validation of the algorithms and procedures used to convert large streams of raw GPS data available from the American Transportation Research Institute into a database of truck trips for use in analysis and modeling of statewide freight truck travel. Also presented are the results from the conversion of 4 months of raw GPS data from Florida, totaling more than 145 million GPS records, into a database of more than 1.2 million truck trips within, into, and out of Florida. The procedures and implementation details described in this paper will be useful to transportation planning agencies in converting raw GPS data streams into more usable truck trip databases for truck flow modeling and analysis at statewide and megaregional levels.